« first day (3221 days earlier)      last day (1691 days later) » 

3:41 AM
@enumaris Without looking at the vixra link, my guess is that the -1/12 comes from the Ramanujan summation of 1+2+3+4+...
 
4:17 AM
@Slereah Just passed by this on reddit
 
0
Q: Strange account suspension date

G. SmithWhy is this account shown as suspended until the nonsensical date of “Mar 18 '92 at 16:28“? I am not asking why the account is suspended, but why the “until” date is 27 years in the past.

 
4:33 AM
@PM2Ring that would be my guess as well
 
 
3 hours later…
7:29 AM
@JohnRennie Alright I'm here.
@SirCumference Just saw the picture. That's pretty funny :)
 
@ScientistSmithYT hi
 
@JohnRennie Hi
 
@ScientistSmithYT where did we get to?
 
@JohnRennie I don't exactly remember. But we were talking about the old saying of 1 gauss equals 1 photon per unit area. Then went onto the tesla part.
Then we went a little further than that but not by much.
I literally remembered it just a few seconds ago. But I just can't remember.
Oh yes we were talking about virtual photons and I asked what they were.
 
7:46 AM
@ScientistSmithYT Ok, this is going to get a bit weird! :-)
 
@JohnRennie I'm with ya
 
Our current best theory for describing what particles are is quantum field theory (QFT).
 
I've heard of it
 
This assumes there is a field for every type of particle. The articles are excitations of this field. For example there is an electron field that describes electrons. Add energy to this field and you can excite it, and that excited state appears as an electron.
Alternatively take energy out to de-excite the field and an electron disappears.
 
Hmmm. Following.
 
7:49 AM
This approach makes a lot of weird things immediately obvious. For example it explains why all electrons are identical, because they are all just excitations of the same field.
Also it neatly explains how particles are created in colliders like the LHC. The kinetic energy of the incoming particles gets transferred into the quantum field where it appears as new particles.
 
Oh! That's! ... That just helped me figure out something else I've been looking at. Sorry continue.
 
It also explains how particles and antiparticles annihilate. For example when an electron and positron meet the energy in the electron/positron field is transferred to the photon field so the electron and positron disappear and two photons appear.
 
What is meant by quantum field in this case?
 
That's a good question and the answer is that no-one knows what a quantum field actually represents physically. The maths works i.e. when we use QFT to do calculations we get the correct answers.
 
Huh. An area that isn't very well known. Sounds like a job I want to take.
 
7:53 AM
But it isn't clear what the reality of quantum fields is, and indeed it isn't even clear if it makes sense to ask what the reality of quantum fields is.
 
@ScientistSmithYT physicists and philosophers have been arguing about this for decades. The fact no firm conclusions have been reached suggests that either it's a very hard question or that the question doesn't make sense.
Anyhow, the point of all this was to try and explain what virtual particles are ...
 
What if we are asking the question slightly wrong? What if instead of asking about the particles we ask about the reocuring effects we see?
 
@ScientistSmithYT that too has been argued about for decades. Most of us just accept ti works and get on with life.
 
This seems like a very hard task that wont be solved anytime soon. But yet I think the magnetic field properties are hiding something of greatness from us. I have been studying this particular stuff all of my life pretty much and I know a lot but not enough to even come close to solving it.
 
7:58 AM
Virtual particles?
 
What if another state of matter is found and we had no clue this whole time. And yes virtual particles but I called it invisible photons until now.
 
No, I mean can we leave the metaphysics aside and concentrate on explaining what virtual particles are?
 
I've spent 16 years of my life studying this as closely as I possibly can.
Oh yes, sorry go ahead
 
OK. Quantum fields can have many different types of excitation. The excitations that represent the particles are the states called Fock states. These basically look like infinite plane waves.
If you remember back to when you learned the Schrodinger equation you found the wavefunction of a free particle was an infinite plane wave.
Well the Fock states are the QFT equivalent of this.
But ...
Quantum fields can have states that aren't Fock states and don't look like particles.
 
Wait what? How? Huh?
Isn't that very rare?
 
8:03 AM
@ScientistSmithYT quite the opposite. The vast majority of states in interacting QFTs are not Fock states.
These don't have any simple interpretation, but we can approximate them as a combination of Fock states i.e. we can describe them as if they were made up from a sum of particles.
 
Hi john
 
And the imaginary particles we use in this sum are the virtual particles.
 
So inside the larger particle there are many more?
 
So the bottom line is that virtual particles are a mathematical device for approximating the states of quantum fields that aren't simple Fock states.
 
How many more particle are inside 1 Fock state?
 
8:05 AM
I must emphasise this - they are a mathematical device. Virtual particles are not actually particles.
 
Oh.
I must be a little confused then.
So they are just formulas?
 
Now, we can describe a magnetic field using virtual photons, and this works, but this does not mean the field is actually made of photons. It is not.
The virtual photons are a mathematical device used to describe the field but they do not actually exist.
@yuvrajsingh hi
 
Then what is the magnetic field made of?
 
It is not made of anything.
 
What? That's impossible?.. Right?
 
8:09 AM
Quantum field theory describes a magnetic field as interactions between particles mediated by the quantum field.
 
So the magnetic field is an effect created by other interactions occurring?
By interactions I'm referring to the air surrounding the magnet.
 
I don't think it's possible to say much more at this level. The magnetic field is one of the phenomena we get when we quantise the electromagnetic four-vector. It turns out to be a far more complicated object than we all learned about as undergrads.
But the bottom line is IT ISN'T MADE OF PHOTONS :-)
 
Ok got it.
Its super late for me anyways. I've had a super long day. Go up at 6 this morning and now its 2 in the morning the next day.
Thanks for the insight. I've learned a lot of things. And its changed a lot of my thinking now.
It's time for me to get some sleep any ways. Night everyone. Thanks again, this has changed a lot of what I knew. @JohnRennie
 
8:30 AM
@JohnRennie Well I mean
It can be described that way I think?
 
@Slereah you can describe it as a sum of virtual photon states, but that doesn't mean the virtual photon states are real physical objects.
 
@JohnRennie I mean it's the whole epistemology thing
You could say none of the objects are real
Physical models aren't about realness
 
8:47 AM
in the end we only get measurements on the other hand
 
On the plus side the debate has given generations of philosophers something to do in between gazing out of the window.
 
9:22 AM
Hey guys, would an arbitrary macrostate of the universe have a Poincaré recurrence time or some analog of it? Could it be proven that such a term would need to be finite or infinite?
 
@JohnRennie Tell me about it
@Skyler Poincaré recurrence times are tricky with realistic models of the universe
There is a theorem stating that some FRWL models of the universe admit no recurrence time
 
So in that metric would the recurrence time be infinity, or would the concept of a recurrence time not exist (some kind of true divide by zero unrenormalizable /undefined error)
 
Well it would not exist, yes
It's an old theorem of Tipler if you're interested
 
9:50 AM
if e = m * c^2, how come can light have energy, but no mass?
 
@yasar the full equation is:
$$ E^2 = p^2c^2 + m^2c^4 $$
where $p$ is the momentum.
For light $m = 0$ so the equation reduces to $E = pc$
Photons can have energy because they have a momentum even though they have no mass.
 
So, they are more like travelling energy
 
45
Q: Do photons truly exist in a physical sense or are they just a useful concept like $i = \sqrt{-1}$?

JensReading about photons I hear different explanations like "elementary particle", "probability cloud", "energy quanta" and so forth. Since probably no one has ever seen a photon (if "seen" it supposedly - and rather conveniently - ceases to exist) but many experiments seem to verify its properties ...

 
Similarly, photons can have momentum because they have energy, even though they have no mass. ;)
 
10:06 AM
Not that we physicists use circular arguments :-)
 
@JohnRennie That article poses another interesting question, how come they interact with gravity without no mass :D
If things with no mass interact with gravity, what does it mean to have a mass anymore :/
 
10
Q: How can gravity affect light?

math and mountainsI understand that a black hole bends the fabric of space time to a point that no object can escape. I understand that light travels in a straight line along spacetime unless distorted by gravity. If spacetime is being curved by gravity then light should follow that bend in spacetime. In Newton...

30
Q: How is light affected by gravity?

PriestVallonLight is clearly affected by gravity, just think about a black hole, but light supposedly has no mass and gravity only affects objects with mass. On the other hand, if light does have mass then doesn't mass become infinitely larger the closer to the speed of light an object travels. So this woul...

General relativity gives the simplest explanation, though that's not much use unless you understand general relativity :-)
 
What I read from all of this is that, mass is either an imaginary unit or emergent property of energy
 
I'd prefer to say that mass is a kind of energy.
 
@yasar none of this is correct
 
10:14 AM
Just as an entity can have a certain kinetic energy, it can have a certain rest energy, aka mass.
 
they have no rest energy, ergo no rest mass
little different but that makes a big diff in this context
 
 
1 hour later…
11:22 AM
@Skyler Does having no rest mass mean "it cannot be stationary"
 
@yasar yes. More precisely it means it has no rest frame.
 
@JohnRennie by that you mean, according to all observers, a photon must always be moving?
 
@yasar yes. The local speed of a photon is always equal to $c$.
 
 
4 hours later…
3:22 PM
@JohnRennie hi can you define what is difference between enthalpy and heat
When we discuss about gibbs free energy enthalpy given us -surround enthalpy which equal to gibbs free energy this enthalpy is energy,and heat is also a energy than what is difference between them
 
4:11 PM
heat isn't a state function.
 
@yuvrajsingh At constant pressure heat and enthalpy are the same, and in many cases we are working at constant pressure.
This is very often the case in chemistry.
 
@Loong very charitably, you could say that it's the kinetic energy gained by a mega (million) of electrons moving through a potential difference
(but yeah, better to just do MeV)
 
4:41 PM
Dropping a philosophy question here because the symposium is too inactive:
I wonder what does it feels like to be e.g. consciously not thinking about apples...
 
Can't say
I think about apples all the time
.oO(🍏)
2
 
apple
 
@enumaris you've used tensorflow haven't you?
 
Yes I have
quite extensively
 
Can you help with this:
in Problem Solving Strategies, 5 hours ago, by Ajay Mishra
(a,b), (c,d) = mnist.load_data()
Ajay is trying to load data but his PC is behind a proxy server and he has no direct internet access.
Is there a way to tell tensorflow to use a web proxy?
 
4:57 PM
hmmm...well he can always just manually download the data
Arguments:

path: if you do not have the index file locally (at '~/.keras/datasets/' + path), it will be downloaded to this location.
 
No way to use a web proxy then?
 
so you can download manually to a folder and then call mnist.load_data('path\to\folder')
not from that keras API I don't think
 
Oh well, thanks. I did try Googling it and found some suggestions you can use environment variables but it wasn't clear exactly how.
 
If you want to load natively in tensorflow rather than keras you can use this method: tensorflow.org/datasets/api_docs/python/tfds/load
but I don't know if that one allows using a web proxy either
Deep learning frameworks are still relatively new so there may be a lot of features not implemented yet
 
5:19 PM
a phrase which I seem to be butting up against lately: reproducing kernel hilbert space
 
what's it mean
 
i dunno. i've just been seeing it in bits and pieces lately
when people talk about applications of hilbert space to prob/stats
 
uh...
 
6:10 PM
Got it figured out:
Some long time ago in the past, I was reading some newspaper and saw a figure that my friends next to me think it is a face
But I don't want to see it as a face, thus I focus on specific features of that figure and shortly after, I ceased to see it as a face
It's actually pretty hard because consciously speaking, it tries so hard to snap back into seeing a face
and with that, a proposal to win The Game:
Whenever you realise you thought of The Game, use the above procedures to non think The Game, then The Game ends up never thought of, and you can avoid losing The Game
 
wat
 
Take this figure as an example, I see it as a meh emoticon
But I do not want to see it as a meh emoticon, thus I tried to think of something that is similar in appearance but not whatever I think I am seeing right now (a meh emoticon), At this point in time, I should be not thinking the meh emoticon, since I am only focus on not seeing what I am currently seeing, but not directly the meh emoticon impression
The think I thought of is a power socket.
Then slowly, the three horizontal lines become identified with the concept of the holes in the power socket, and the meh emoticon perception bec
Now the joke thought:
> 0. There is a game (consequence of zeroth law of thermodynamics)
1. You can't win. (consequence of first law of thermodynamics)
2. You can't break even. (consequence of second law of thermodynamics)
3. You can't even get out of the game. (consequence of third law of thermodynamics)
If I can somehow replicate this non thinking strategy somehow at the physical level, it should allow me to break the 1st law of thermodynamics
(obviously a joke since you cannot break the 1st law of thermodynamics, except in GR where global energy is not well defined)
 
 
1 hour later…
7:21 PM
uh...
 
7:35 PM
lol
apparently RKH spaces are significant in (some branch) of machine learning? (e.g. the slides here: gatsby.ucl.ac.uk/~gretton/coursefiles/Slides4A.pdf)
 
oh
glancing over it it seems they are doing something like the kernel trick used in training SVMs to give non-linear decision boundaries
 
right
 
how is it that kittens have sharper claws that any known substance?!
I theorise they are actually made of something subatomic, which litterally gets in between the atoms of my foot when they grab on
 
Murphy's law as applied to kittens: anything you like which they can scratch, they will scratch
@enumaris a bit of elaboration on that: stats.stackexchange.com/a/233391/211265
 
7:51 PM
mmmhm...
 
it perks my curiousity because 1) it's hilbert space applied to stats, which i'm interested in lately, and 2) the reproducing condition is the same one you see in QM for the feynman propagator
or it would be, if that were more than an abstract...
 
o.o
 
weirdly, inspire.hep has pics from it: inspirehep.net/record/1645695?ln=en
 
I think in ML the idea is you make your training data your feature anchors which means you're working in an extremely high dimensional space
using the kernel trick allows you to not have to actually transform data into this extremely high dimensional space in order to perform training
 
7:55 PM
I haven't had to use SVMs and kernels all that much though...they don't really appear in deep learning
 
ah, okay
 
Andrew Ng's ML course has a short class on this
 
makes sense. there's a reason I covered my butt by saying "(some branch) of machine learning"
 
8:09 PM
Hello,

Could someone re-explain this "terrible" Hamiltonian, as they say, I couldn't recognize why?
$\begin{aligned} H &=\int \frac{d^{3} p}{(2 \pi)^{3}} E_{\vec{p}}\left(b_{\vec{p}}^{s \dagger} b_{\vec{p}}^{s}-c_{\vec{p}}^{s} c_{\vec{p}}^{s \dagger}\right) \\ &=\int \frac{d^{3} p}{(2 \pi)^{3}} E_{\vec{p}}\left(b_{\vec{p}}^{s \dagger} b_{\vec{p}}^{s}-c_{\vec{p}}^{s \dagger} c_{\vec{p}}^{s}+(2 \pi)^{3} \delta^{(3)}(0)\right) \end{aligned}$
Naturally, the first line tells us, by annihilating a particle with momentum $p$ leads to the creation of an antiparticle with the same momentum.
i.e. why we try to make contradiction by substituting the second operator using the corresponding commutation relation?
then we say the Hamiltonian is not lower bound?
 
seems like you could drop $b$ out of there entirely
(in the sense that, if it's not bounded below, it's because of the c's and not the b's)
 
Why it's because of the c's and not the b's?
Or you want to say, by logical reasoning, the total energy is the sum of those for particles and antiparticles?
and not the substruction?
 
 
1 hour later…
9:29 PM
is it possible to flux pin a sufficiently large superconducting disk to earth's magnetic field?
 
 
2 hours later…
11:13 PM
@Student404Mus The $-E_{\vec{p}} c_{\vec{p}}^{s \dagger} c_{\vec{p}}^s$ term means the more particles created by $c_{\vec{p}}^{s \dagger}$ in the theory there are, the lower the energy of the system gets, i.e. negative energy potentially even going to minus $\infty$ negative energy, which is nonsense
 

« first day (3221 days earlier)      last day (1691 days later) »